Prospective data on the relationship between skin test anergy, PPD tuberculin positivity and active tuberculosis (TB) disease and HIV infection in intravenous drug users (IVDUs) are sparse. In addition, the compliance level, and side-effects associated with isoniazid TB preventive therapy in HIV seropositive IVDUs are not known. To address these issues we propose to conduct a two part study over 5 years in a cohort of 600 HIV-1 seropositive and 600 HIV-1 seronegative IVDUs, who are almost exclusively not in drug treatment programs, in Baltimore City. The first part of the study involves screening the cohort for asymptomatic TB infection with PPD tuberculin skin tests. Eligibility for chemoprophylaxis will be defined by a tuberculin reaction of equal to or greater than 10 mm. diameter induration after 48 hours in HIV-1 seronegatives, and equal to or greater than 5 mm. diameter induration in HIV-1 seropositives. Participants will also be skin tested to determine if they are anergic (i.e., unable to mount a cell-mediated hypersensitivity reaction to skin- test antigens) to allow proper interpretation of negative PPD tests. HIV seropositive participants who are anergic will also be offered chemoprophylaxis in accordance with new CDC guidelines. Participants will receive tuberculin and anergy tests at baseline, and will be retested every 6 months. This will allow us to describe the natural history of development of anergy in HIV seropositives, tuberculin boosting with repeated testing, and incidence of tuberculin conversions by HIV serostatus. The second part of the study will determine relative risk for developing active TB disease in the cohort. Risk will be calculated depending on prior diameter induration from tuberculin testing, HIV status, CD4 cell count, compliance with isoniazid chemoprophylaxis, presence or absence of adverse reactions to isoniazid, drug use patterns, demographic variables and markers of poverty. In particular, risk of TB disease will be determined in anergic HIV positive participants. Since it is unethical to withhold isoniazid from a comparison group, effectiveness of isoniazid be indirectly assessed by comparing TB disease rates in compliant and non-compliant participants (adjusting for confounding and bias).